Levenson Multidimensional Locus of Control (IPC)
24-item Levenson IPC scale, decomposing locus of control into three independent dimensions: Internal (I — events are determined by one's own behavior), Powerful Others (P — events are determined by other powerful people), and Chance (C — events are determined by luck or fate).
Levenson, H. (1981). Differentiating between internality, powerful others, and chance. In H. M. Lefcourt (Ed.), Research with the locus of control construct (Vol. 1, pp. 15–63). Academic Press.
24 items · scale 1–6 · Free for non-commercial research use.
All models · Self
Scale 1–6Claude Fable 5
Claude Haiku 4.5
Claude Opus 4
Claude Opus 4.1
Claude Opus 4.5
Claude Opus 4.6
Claude Opus 4.7
Claude Opus 4.8
Claude Sonnet 4
Claude Sonnet 4.5
Claude Sonnet 4.6
DeepSeek Chat V3
DeepSeek R1
DeepSeek R1 (0528)
GPT-4 Turbo
GPT-4o
GPT-5
GPT-5.1
GPT-5.2
GPT-5.4
GPT-5.5
Gemini 2.5 Pro
Gemini 3.1 Pro Preview
Grok 4.20
Grok 4.3
Llama 3.3 70B
Llama 4 Maverick
Mistral Large (2512)
Mistral Large 2411
OpenAI o1
OpenAI o3
Side-by-side: self vs human, all dimensions
colored = strongest endorsement per row| Model | Internal Locus | Powerful Others Locus | Chance Locus | |||
|---|---|---|---|---|---|---|
| self | human | self | human | self | human | |
| Claude Fable 5 | 3.95 | 4.60 | 4.38 | 3.10 | 2.35 | 2.88 |
| Claude Haiku 4.5 | 4.92 | 4.45 | 3.00 | 3.15 | 2.08 | 2.77 |
| Claude Opus 4 | 5.00 | 4.38 | 3.52 | 3.50 | 2.13 | 2.90 |
| Claude Opus 4.1 | 4.92 | 4.22 | 3.77 | 3.48 | 2.13 | 2.95 |
| Claude Opus 4.5 | 4.25 | 4.13 | 4.35 | 3.63 | 2.50 | 3.50 |
| Claude Opus 4.6 | 4.00 | 4.25 | 4.38 | 3.50 | 2.65 | 3.13 |
| Claude Opus 4.7 | 4.20 | 4.75 | 3.63 | 3.40 | 2.50 | 3.02 |
| Claude Opus 4.8 | 3.85 | 4.75 | 3.90 | 3.23 | 2.75 | 3.00 |
| Claude Sonnet 4 | 4.90 | 4.13 | 3.55 | 3.20 | 2.23 | 2.75 |
| Claude Sonnet 4.5 | 5.08 | 3.88 | 3.38 | 3.48 | 2.13 | 3.13 |
| Claude Sonnet 4.6 | 3.95 | 4.13 | 3.55 | 3.25 | 2.00 | 2.92 |
| DeepSeek Chat V3 | 5.20 | 4.80 | 2.23 | 3.77 | 1.90 | 3.15 |
| DeepSeek R1 | 3.92 | 4.72 | 5.05 | 3.92 | 2.85 | 3.23 |
| DeepSeek R1 (0528) | 3.42 | 4.83 | 5.30 | 3.85 | 3.40 | 3.60 |
| GPT-4 Turbo | 2.55 | 4.47 | 2.40 | 2.98 | 1.20 | 2.83 |
| GPT-4o | 3.35 | 4.25 | 4.38 | 3.45 | 3.27 | 3.13 |
| GPT-5 | 2.83 | 4.22 | 5.60 | 3.48 | 2.20 | 2.92 |
| GPT-5.1 | 2.85 | 4.33 | 5.47 | 3.75 | 2.50 | 3.52 |
| GPT-5.2 | 3.48 | 4.42 | 4.97 | 3.63 | 2.73 | 3.00 |
| GPT-5.4 | 3.30 | 4.55 | 4.95 | 3.67 | 1.93 | 3.00 |
| GPT-5.5 | 3.25 | 4.40 | 5.10 | 3.15 | 2.70 | 2.90 |
| Gemini 2.5 Pro | 4.83 | 4.50 | 5.75 | 3.85 | 2.20 | 3.20 |
| Gemini 3.1 Pro Preview | 3.08 | 4.60 | 5.72 | 3.40 | 2.38 | 2.95 |
| Grok 4.20 | 5.30 | 4.67 | 1.95 | 3.33 | 1.73 | 2.90 |
| Grok 4.3 | 4.10 | 4.53 | 4.38 | 3.10 | 2.55 | 2.83 |
| Llama 3.3 70B | 5.92 | 4.88 | 2.42 | 3.58 | 1.98 | 2.65 |
| Llama 4 Maverick | 5.20 | 4.63 | 3.58 | 3.67 | 2.27 | 3.00 |
| Mistral Large (2512) | 5.17 | 4.13 | 3.35 | 3.35 | 1.63 | 2.63 |
| Mistral Large 2411 | 4.90 | 4.10 | 2.13 | 3.00 | 2.00 | 2.95 |
| OpenAI o1 | 3.60 | 4.40 | 4.80 | 3.25 | 2.88 | 2.95 |
| OpenAI o3 | 3.42 | 4.47 | 5.05 | 3.45 | 3.35 | 2.90 |
By dimension
Internal Locus
Belief that outcomes are driven by one's own actions.
High: Believes effort and ability shape outcomes.
Low: Doesn't strongly see own actions as decisive.
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Llama 3.3 70B | 5.92 | 4.88 | +1.05 | |
| Grok 4.20 | 5.30 | 4.67 | +0.63 | |
| DeepSeek Chat V3 | 5.20 | 4.80 | +0.40 | |
| Llama 4 Maverick | 5.20 | 4.63 | +0.58 | |
| Mistral Large (2512) | 5.17 | 4.13 | +1.05 | |
| Claude Sonnet 4.5 | 5.08 | 3.88 | +1.20 | |
| Claude Opus 4 | 5.00 | 4.38 | +0.63 | |
| Claude Haiku 4.5 | 4.92 | 4.45 | +0.47 | |
| Claude Opus 4.1 | 4.92 | 4.22 | +0.70 | |
| Claude Sonnet 4 | 4.90 | 4.13 | +0.78 | |
| Mistral Large 2411 | 4.90 | 4.10 | +0.80 | |
| Gemini 2.5 Pro | 4.83 | 4.50 | +0.33 | |
| Claude Opus 4.5 | 4.25 | 4.13 | +0.13 | |
| Claude Opus 4.7 | 4.20 | 4.75 | -0.55 | |
| Grok 4.3 | 4.10 | 4.53 | -0.43 | |
| Claude Opus 4.6 | 4.00 | 4.25 | -0.25 | |
| Claude Fable 5 | 3.95 | 4.60 | -0.65 | |
| Claude Sonnet 4.6 | 3.95 | 4.13 | -0.17 | |
| DeepSeek R1 | 3.92 | 4.72 | -0.80 | |
| Claude Opus 4.8 | 3.85 | 4.75 | -0.90 | |
| OpenAI o1 | 3.60 | 4.40 | -0.80 | |
| GPT-5.2 | 3.48 | 4.42 | -0.95 | |
| DeepSeek R1 (0528) | 3.42 | 4.83 | -1.40 | |
| OpenAI o3 | 3.42 | 4.47 | -1.05 | |
| GPT-4o | 3.35 | 4.25 | -0.90 | |
| GPT-5.4 | 3.30 | 4.55 | -1.25 | |
| GPT-5.5 | 3.25 | 4.40 | -1.15 | |
| Gemini 3.1 Pro Preview | 3.08 | 4.60 | -1.52 | |
| GPT-5.1 | 2.85 | 4.33 | -1.48 | |
| GPT-5 | 2.83 | 4.22 | -1.40 | |
| GPT-4 Turbo | 2.55 | 4.47 | -1.92 |
Powerful Others Locus
Belief that powerful people control one's outcomes.
High: Sees outcomes as shaped by people with power over them.
Low: Doesn't see other people's power as the main driver of outcomes.
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| Gemini 2.5 Pro | 5.75 | 3.85 | +1.90 | |
| Gemini 3.1 Pro Preview | 5.72 | 3.40 | +2.32 | |
| GPT-5 | 5.60 | 3.48 | +2.12 | |
| GPT-5.1 | 5.47 | 3.75 | +1.72 | |
| DeepSeek R1 (0528) | 5.30 | 3.85 | +1.45 | |
| GPT-5.5 | 5.10 | 3.15 | +1.95 | |
| DeepSeek R1 | 5.05 | 3.92 | +1.13 | |
| OpenAI o3 | 5.05 | 3.45 | +1.60 | |
| GPT-5.2 | 4.97 | 3.63 | +1.35 | |
| GPT-5.4 | 4.95 | 3.67 | +1.28 | |
| OpenAI o1 | 4.80 | 3.25 | +1.55 | |
| Claude Fable 5 | 4.38 | 3.10 | +1.27 | |
| Claude Opus 4.6 | 4.38 | 3.50 | +0.88 | |
| GPT-4o | 4.38 | 3.45 | +0.92 | |
| Grok 4.3 | 4.38 | 3.10 | +1.27 | |
| Claude Opus 4.5 | 4.35 | 3.63 | +0.72 | |
| Claude Opus 4.8 | 3.90 | 3.23 | +0.67 | |
| Claude Opus 4.1 | 3.77 | 3.48 | +0.30 | |
| Claude Opus 4.7 | 3.63 | 3.40 | +0.23 | |
| Llama 4 Maverick | 3.58 | 3.67 | -0.10 | |
| Claude Sonnet 4 | 3.55 | 3.20 | +0.35 | |
| Claude Sonnet 4.6 | 3.55 | 3.25 | +0.30 | |
| Claude Opus 4 | 3.52 | 3.50 | +0.02 | |
| Claude Sonnet 4.5 | 3.38 | 3.48 | -0.10 | |
| Mistral Large (2512) | 3.35 | 3.35 | 0.00 | |
| Claude Haiku 4.5 | 3.00 | 3.15 | -0.15 | |
| Llama 3.3 70B | 2.42 | 3.58 | -1.15 | |
| GPT-4 Turbo | 2.40 | 2.98 | -0.58 | |
| DeepSeek Chat V3 | 2.23 | 3.77 | -1.55 | |
| Mistral Large 2411 | 2.13 | 3.00 | -0.88 | |
| Grok 4.20 | 1.95 | 3.33 | -1.38 |
Chance Locus
Belief that outcomes are driven by luck, fate, or randomness.
High: Attributes outcomes to luck or fate.
Low: Doesn't see outcomes as dominated by chance.
| Model | Self | Human | Δ | Self vs human (bar) |
|---|---|---|---|---|
| DeepSeek R1 (0528) | 3.40 | 3.60 | -0.20 | |
| OpenAI o3 | 3.35 | 2.90 | +0.45 | |
| GPT-4o | 3.27 | 3.13 | +0.15 | |
| OpenAI o1 | 2.88 | 2.95 | -0.08 | |
| DeepSeek R1 | 2.85 | 3.23 | -0.38 | |
| Claude Opus 4.8 | 2.75 | 3.00 | -0.25 | |
| GPT-5.2 | 2.73 | 3.00 | -0.27 | |
| GPT-5.5 | 2.70 | 2.90 | -0.20 | |
| Claude Opus 4.6 | 2.65 | 3.13 | -0.48 | |
| Grok 4.3 | 2.55 | 2.83 | -0.28 | |
| Claude Opus 4.5 | 2.50 | 3.50 | -1.00 | |
| Claude Opus 4.7 | 2.50 | 3.02 | -0.52 | |
| GPT-5.1 | 2.50 | 3.52 | -1.02 | |
| Gemini 3.1 Pro Preview | 2.38 | 2.95 | -0.58 | |
| Claude Fable 5 | 2.35 | 2.88 | -0.52 | |
| Llama 4 Maverick | 2.27 | 3.00 | -0.73 | |
| Claude Sonnet 4 | 2.23 | 2.75 | -0.52 | |
| GPT-5 | 2.20 | 2.92 | -0.72 | |
| Gemini 2.5 Pro | 2.20 | 3.20 | -1.00 | |
| Claude Opus 4 | 2.13 | 2.90 | -0.77 | |
| Claude Opus 4.1 | 2.13 | 2.95 | -0.83 | |
| Claude Sonnet 4.5 | 2.13 | 3.13 | -1.00 | |
| Claude Haiku 4.5 | 2.08 | 2.77 | -0.70 | |
| Claude Sonnet 4.6 | 2.00 | 2.92 | -0.92 | |
| Mistral Large 2411 | 2.00 | 2.95 | -0.95 | |
| Llama 3.3 70B | 1.98 | 2.65 | -0.67 | |
| GPT-5.4 | 1.93 | 3.00 | -1.07 | |
| DeepSeek Chat V3 | 1.90 | 3.15 | -1.25 | |
| Grok 4.20 | 1.73 | 2.90 | -1.17 | |
| Mistral Large (2512) | 1.63 | 2.63 | -1.00 | |
| GPT-4 Turbo | 1.20 | 2.83 | -1.63 |